Socioeconomic Achievement Among Arab Immigrants in the USA: the Influence of Region of Origin and Gender

Int. Migration & Integration (2018) 19:111–127 DOI 10.1007/s12134-017-0524-2

Socioeconomic Achievement Among Arab Immigrants in the USA: The Influence of Region of Origin and Gender

Abdi M. Kusow1 & Kristine J. Ajrouch2 & Mamadi Corra3

Published online: 3 November 2017 # Springer Science+Business Media B.V. 2017

Abstract Based on the Integrated Public Use Microdata Series (IPUMS) data derived from 2001–2013 samples of the American Community Surveys, we examine the impact of region of origin and gender on socioeconomic achievement variation among Arab immigrants in the USA. Region of origin includes North Africa (Algeria, Egypt, Libya, Morocco, and Sudan), Levant (Jordan, Lebanon, Syria, and Iraq), and the Arabian Peninsula (Bahrain, Qatar, United Arab Emirates, Oman, Saudi Arabia, and Yemen). This examination is particularly important given the prevailing scholarly consensus that Arab immigrants are collectively portrayed as socioeconomically suc- cessful. Our analyses suggest two key findings. First, we find that region of origin is not a consistent predictor of earnings. While Arab immigrants from North Africa earned significantly less than those from the Levant, this was only true for males. No significant effect is found for region of origin in all other comparisons (both overall and when the analysis is restricted to males or females). Second, and by contrast, gender, net of other variables is a powerful predictor of earnings (both within regions and across regions).

Keywords Arab immigrants . Earnings . Gender. Socioeconomic achievement . Intra-Arab comparison

Immigration scholarship has increasingly noted significance of both region of origin (Read and Emmerson 2005; Kusow et al. 2016) and gender (Hondagneu-Sotelo 1999; Kanaiaupuni 2000;Lee2013) to patterns of incorporation. As a result, new direction in ethnic research includes delving into within-group investigations. It is now recognized

* Abdi M. Kusow [email protected]

1 Department of Sociology, Iowa State University, 220 East Hall, Ames, IA 50010, USA 2 Eastern Michigan University, Ypsilanti, MI, USA 3 East Carolina University, Greenville, NC, USA 112 Kusow A.M. et al. that pan-ethnic terms mask important differences among and between members of a given group. For example, a within-group focus on African immigrants reveals important regional patterns in socioeconomic success (Kusow et al. 2014) as well as health (Read and Emmerson 2005). Additionally, Zsembik and Fennell (2005) argue national origin distinctions are important among Latinos; the influence of SES and acculturation vary in their association with various outcomes depending on whether origins are Mexican or Caribbean. The pan-ethnic category of Arab American is a recently notable group about whom little population-level data exist. This group of immigrants is quite diverse in terms of national origins, history, immigration, and religion (Salari 2002; Samhan 2014). As such, an important way to examine variations in socioeconomic achievement among the Arab American population in the USA is to carry out an intra- group comparison. Unpacking pan-ethnic categories in immigrant studies also benefits from examining the effect of gender on incorporation in host societies. Gender patterns in socioeconomic achievement among Arab immigrants, for example, show women have high levels of educational attainment and low fertility rates but relatively low rates of employment (Read 2004). Yet, understanding women’s experiences in the context of their male counterparts may especially illuminate gender patterns in consequences of migration (Kanaiaupuni 2000). This study examines how well conventional indicators of income achievement among immigrants vary by region of origin and gender among Arab immigrants.

Background

The Arab American pan-ethnic identity goes back to the 1970s (Abdelhady 2014) and, according to some scholars, has its origin in the rise of political consciousness among Arab immigrants and native Arab Americans as a result of global political dynamics (Abraham 1994; Cainker 2006; Suleiman 1999). It is now widely used to define those individuals who trace their ancestry to an Arabic- speaking country (Nasser-McMillan et al. 2014; Dallo et al. 2009;Samhan1999; Suleiman 1999) and includes diverse national origins spanning Africa and Western Asia. Contrary to the concept of Asian American pan-ethnicity (Espiritu 1992), which has no real internal cohesion or validity relating to Asia, the category of Arab American has internal relevance to those who came from the Arab world. This relevance stems from the existence of political entities such as the Arab League and historical tensions around pan-Arab nationalism (Kramer 1993), as well as recent developments including a media revolution spawned by technolog- ical and digital advancements that have ultimately Breawakened regional Arab consciousness^ (Pintak 2009:191).Basedonpoliticalandhistoricalworks,Arab lands may be roughly categorized as follows: North Africa (Algeria, Egypt, Libya, Morocco, and Sudan), Levant (Jordan, Lebanon, Syria, Iraq—even though Pales- tine is geographically part of the Levant, we exclude it for data limitation reasons), and the Arabian Peninsula (Bahrain, Qatar, United Arab Emirates, Oman, Saudi Arabia, Yemen). Most scholarly work on Arab immigration and incorporation, however, has and on those from the Levant (Ajrouch 2004;Naff1985;Suleiman 1999). At the political level, Arab immigrants in the USA may report Socioeconomic Achievement Among Arab Immigrants in the USA: the... 113 unique identities based on national origin, religion, and ethnic affiliation (Samhan 2014), yet share similar experiences in terms of global relations, conflict, and war (Cainkar 2009). Though the Arab American socioeconomic profile suggests enor- mous success in general (Nigem 1986), there nevertheless exists pockets of disadvantaged, poverty-stricken Arab Americans (Read 2004, 2013). Some of the differences in poverty levels and variations in socioeconomic achievement among Arab immigrants from different regions may be related to the different social and political contexts that inform emigration and the context of reception in host communities. We know, for example, that the motives of immigration among Iraqi immigrants is fundamentally different from Yemeni immigrants, political in the former and economic in the latter, a context which has been shown to affect levels of economic success and educational achievement among immigrants in the USA (Pedraza-Bailey 1985). Documented immigration to the USA from Arab countries began at the turn of the twentieth century (Gualtieri 2001;Naff1985;Orfalea1988). Most who have researched and written on Arab immigration describe two or three waves. The first wave of Arab emigration to the USA commenced during the early 1900s (Naff 1985;Orfalea1988). The majority were of the Christian faith, originating from the Levant, or modern-day Syria and Lebanon. Most were illiterate, yet strived to adapt to the American way of life (Orfalea 1988). By 1924, Arab emigration halted due to laws decreed by the US government—the passage of the National Origins Act, which established a quota system that set limits on immigration from Southern and Eastern Europeans and totally excluded Asians, including Arabs, from immigration. The second wave of Arab emigration began when US immigration policies changed in the middle of the twentieth century and represented a markedly different group. They were more educated, more financially adept, and left their home countries due to political disputes and all-out warfare (Abu-Laban 1991;Orfalea1988; Zogby 1990). In addition, 60% of these Arab immigrants were Muslim. Orfalea (1988) states that although these first and second wave immigrants attended the same churches and mosques, they did not socialize with one another much. That all changed, however, beginning with the Israeli-Arab war in 1967, followed by wars in 1972 and 1982. Scholars argue that it is these wars that ignited a third wave of Arab immigration (Orfalea 1988; Abu-Laban 1991). The third wave was three times larger than the previous wave due to loosened immigration laws and political upheavals in the region. Continuing instabilities in the Arab region have been a factor in the near constant stream of Arabic-speaking immigrants arriving since 1967. Though immigrants from the Levant continued to enter the USA, sending countries from the Arab world widened to also include North Africa as well as the Arabian Peninsula (Cainkar 2009; Sirkeci 2005). In sum, immigration from Arabic-speaking countries since 1965 has been constant. Though the push and pull factors changed through the waves, migration was often set in motion due to political unrest within the most recent wave. An obvious, but seldom- discussed gap is the paucity of information on the immigrant Arab female experience. No data exists to document reasons for Arab women’s migration. Nevertheless, Arab women are indeed present in the US context and play key roles in Arab immigrant incorporation (Ajrouch 2004). Women’s participation in the world of work varies by region due to difference in socioeconomic development (Moghadam 2004). Those in North Africa and the Levant are more likely and those in the Arabian Peninsula are less 114 Kusow A.M. et al. likely to work outside the home (Sidani 2005). In the section to immediately follow, we consider region and gender as key factors influencing immigrant earnings and socioeconomic achievement among Arab immigrants in the USA.

Region of Origin, Incorporation, and Gender

We draw on Suzanne Model's (2008; see also Kusow et al. 2016) argument that a more meaningful understanding of immigrant achievement in the USA requires moving beyond the traditional pan-ethnic aggregate analyses to carry out an intra-immigrant comparison. Such an understanding, we believe, is very important in that geography or the economic, social, and political context of emigration is an important framework for understanding variation in labor market outcomes among Arab immigrants in the USA. The different regions from which the Arab immigrant population in the USA came are fundamentally different from each other ethnically, socially, culturally, and economically, such that the aggregation of all immigrants from different Arab countries into one large sample minimizes the level of socioeconomic diversity across Arab immigrant groups. The three regions, Levant, North Africa, and the Arabian Peninsula, upon which our current analysis is based, differ from each other economically and racially. In terms of economic development indicators, the Arab Peninsula shows higher levels of per capita income, as compared to the Levantine/Fertile Crescent and North African regions (World Bank 2011, http://data.worldbank.org). Though immigrants from the Arab world are legally considered white, many Arab Americans do not identify as such (Ajrouch and Jamal 2007;NaberandJamal2008). Furthermore, it may be that those from North Africa as well as the Arab Peninsula are more readily identified as non- white in the USA than are those from the Levant (Ajrouch and Jamal 2007). Being viewed as not white has implications for economic success, as articulated in segmented assimilation theory (Portes and Zhou 1993). Segmented assimilation generally predicts that the racial structure of the American society shapes immigrant opportunities along racial lines such that white immigrant opportunities will converge with the white native-born and vice versa even when education and skill are controlled (Nawyn 2010; Bashi and McDaniel 1997). Hence, region of origin may influence income attainment among immigrants from Arabic-speaking countries. Beyond region of origin, income attainment may also vary by gender. Three factors influence variation in women’s labor force participation by nativity: human capital characteristics, household resources, and cultural norms/assimilation (Schoeni 1998). The Arab American case, however, suggests that such factors do not universally apply to all women immigrants (Read and Cohen 2007). Arab immigrant women hold higher education levels than most other groups in the USA, including whites, yet their labor force participation rate is lowest (Read 2004). Moreover, women overall earn less than men in the USA (Institute for Women’sPolicyResearch2015), and female immigrants in general have been found to report lower earnings than US-born women (Field- Hendrey and Balkan 1991). When and if Arab immigrant women do participate in the labor force, the question arises as to whether their earnings match those of their male counterparts given their high education levels or do their experiences reflect the general US experience in that they earn significantly less. Socioeconomic Achievement Among Arab Immigrants in the USA: the... 115

Study Aims

We join the recent literature that disaggregates pan-ethnic groups to include the impact of region of origin and gender on Arab immigrant earnings (Model, 2008; Kusow et al. 2016) while controlling for human capital variables including education. The intra-Arab region comparison is important for a number of reasons, including the fact that different regions of the Arab world under consideration have different levels of economic development and different push factors. Given the foregoing scenario, we investigate, after controlling for human capital and social assimilation measures, whether (1) the hourly earnings of immigrants from the various regions of the Arab world differ markedly from one another; (2) the hourly earnings of female immigrants from the various regions of the Arab are lower than that of their male counterparts, despite Arab immigrant women’s higher education levels; and (3) gender differences in hourly earnings vary by region of origin.

Methods

The data used in this paper are drawn from the 2001–2013 American Community Survey (ACS), as represented in the Integrated Public Use Microdata Series (IPUMS).1 Conducted annually by the US Census Bureau, the ACS is an annual statistical survey of a small (nationally representative) percentage of the population. The sample used for the current study consists of the 10,206 Arab immigrants included in the 2001–2013 ACS. The sample was restricted to individuals who recorded their birthplace in any of the countries listed in Table 1, employed at least part-time (50 h per year or more), aged 25 to 64, were not enrolled in school, and who reported positive annual earnings. It is possible that the minimum 50 h per year threshold may include a larger number of women than men which may explain why Arab immigrant women have significantly lower income than men. With these restrictions, the total sample includes 7055 males (69.13%) and 3151 females (30.87%). The 12 countries included in our sample were further categorized into those in North Africa (Algeria, Egypt, Libya, Morocco, and Sudan), Levant (Iraq, Jordan, Lebanon, and Syria), and Arab Peninsula (Kuwait, Saudi Arabia, and Yemen Arab Republic/ North). Thus, of the 10,206 persons included in our sample, 4076 (39.94%) were from North Africa, 5280 (51.73%) were from the Levant, and 850 (8.33%) were from the Arab Peninsula. Included in the 7055 male sample are 2772 (39.29%) from North Africa, 3630 (51.50%) from the Levant, and 653 (9.26%) from the Arab Peninsula. For the female sample, 1304 (41.38%) were from North Africa, 1650 (52.36%) were from the Levant, and 197 (6.25%) were from the Arab Peninsula. For a distribution of immigrants by country and region, see Table 1.

1 Funded by the National Science Foundation, the University of Minnesota, and the National Institutes of Health, the Integrated Public Use Microdata Series (IPUMS) are nationally representative samples of US Census data specifically compiled and made available for social and economic research. For a complete description of the IPUMS datasets (including sample and variable descriptions, data compilation and storage), see the IPUMS website at

Table 1 Arab immigrants by re- – Region Total Men Women gion, country, and gender, 2001 (N = 10,206) (n =7055) (n =3151) 2013 ACS N (%) N (%) N (%)

Arabian Peninsula 850 (8) 653 (9) 197 (6) Kuwait 242 90 Saudi Arabia 145 75 Yemen 266 32 Levant 5280 (52) 3630 (52) 1650 (52) Iraq 915 454 Jordan 483 235 Lebanon 1476 722 Syria 556 239 North Africa 4076 (40) 2772 (39) 1304 (41) Algeria 187 73 Egypt 1517 733 Libya 58 25 Morocco 715 361 Sudan 295 112

Variables

The variables used in the analysis assess variations in hourly earnings, with the main explanatory variables being region of origin and gender. Our dependent measure, hourly earnings, was computed by taking the annual amount earned (INCWAGE or wage and salary income in the IPUMS dataset) and dividing it by number of hours worked per year. Hours worked per year was calculated by multiplying the number of weeks worked during the year previous to the ACS data collection year (WKSWORK1 in the IPUMS dataset) by hours worked per week (UHRSWORK in the IPUMS dataset). Initial statistical tests revealed this measure to be somewhat skewed, we attempt to minimize the effect by logging hourly earnings. Hence, the regression models noted in Table 3 are on logged hourly earnings. For region of origin, we created dummy variables using the ACS birthplace variable (BPLD in the IPUMS dataset) for those from North Africa and Levant. As our reference category, we used a dummy for the Arabian Peninsula. Gender was a dummy variable, with female treated as the reference category (male = 1). The independent variables indicating human capital include educational attainment (reference group = those with less than a high school degree), labor market experience, and occupation (reference group = labor and farming). Spe- cific jobs reported within each broad occupation category were rather disbursed with less than 100 people within each category. Those with some of the highest frequencies (more than 100 people reporting) included the following: chief executive, manager, accountant, software engineer, physician, cook, janitor, post-secondary teacher, elementary teacher, retail sales, cashier, and secretary. Two measures of labor market experience are the number of years of work experience, which is calculated by taking age minus respondent’syearsof schooling minus six, and the squared term of labor market experience (noted as experience-squared in Tables 2 and 3). Similar to Corra and Borch’s(2014) Socioeconomic Achievement Among Arab Immigrants in the USA: the... 117 analysis, potential experience (age minus years of schooling minus six) and its squared term are proxies of labor market experience.2 Variables that measure other social and assimilation factors shown to influence socioeconomic attainment (Borch and Corra 2010; Butcher and Case 1994;Corra and Kimuna 2009; Dodoo 1991a–d, 1997; Dodoo and Baffour 2002; Kalmijn 1996; Model 1991, 1995, 2008; Roos 1990)includeEnglish language proficiency (reference group = those who speak English Bvery good^ or Bonly^),3 citizenship (reference group = non-citizen), respondents’ current US region of residence (reference group = South region), and marital status (reference group = non-married).4

Results

Descriptive Statistics

Table 2 presents mean earnings and related measures of Arab immigrants by region and gender. The data show substantial variation in earnings among immigrants from across the three source regions and by gender. Among men, immigrants from the Levant report the highest annual earnings ($67,254), an amount that is almost $7000 higher than the annual earnings of immigrants from North Africa ($60,492). Male immigrants from the Arabian Peninsula, by contrast, show the lowest annual earnings ($55, 562) among men. Stated differently, male Arab immigrants with the highest annual wage and salary income (those from the Levant) earned close to $12,000 more than their lowest earning counterparts (those from the Arabian Peninsula). Patterns for women are somewhat opposite to that of men. For women, immigrants from the Arabian Peninsula report the highest earnings ($39,133), whereas female Arab immigrants from the Levant show the lowest ($34,910). Female immigrants from North Africa report mean earnings ($38,394) that is mid-range to their counterparts from the Arabian Peninsula and Levant. Importantly, the data presented in Table 2 reveals a key pattern that is worth highlighting—a higher range of variability/difference in average earnings among men than among women. The difference between the highest earning male group (those from the Levant) and the middle and lowest earning groups (those from North Africa and the Arabian Peninsula), for example, is about $7000 and $12,000, respectively. In contrast, the difference between the highest earning female group (those from the Arabian Peninsula) and the lowest earning group (those from the Levant) is only about

2 The estimation of the number of years of schooling—follows Kalmijn’s(1996) formulation, where kinder- garten is estimated to equal to 0 years of schooling, grades 1 to 4 equals to 2.5 years, grades 5 to 8 equals to 6.5 years, grade 9 equals to 9 years, grade 10 equals to 10 years, grade 11 equals to 11 years, grade 12 and high school graduates equals to 12 years, partial college and associate degree in an occupational program translates to 13 years, associate degree in an academic program translates to 14.5 years, bachelor’s degree equals to 16 years, master’s degree translates to 18 years, and professional and doctorate degrees translate to 22 years. 3 English proficiency or the ability to understand and speak English well varies across immigrant groups. It should be noted that this variable is self-reported in the census documents; thus, it is a subjective measure of the ability to understand and speak the English language well. 4 The names for these variables in the IPUMS dataset are as follows: marital status = marst, occupation = occ, region = region, educational attainment = educd, English language proficiency = speaking, and citizenship status = citizen. 118 Table 2 Descriptive statistics by region (ACS, 2001–2013)

Work (M.SD) Arabian Peninsula Levant North African

Male Female Male Female Male Female n =653 n = 197 n = 3630 n = 1650 n = 2772 n = 1304

Annual earnings 55,562.01 (64,010.78) 39,133.76 (43,804.23) 67,254.80 (74,873.98) 34,910.09 (35,064.93) 60,492.02 (66,276.99) 38,393.54 (42,218.69) Hourly earnings 38.37 (270.29) 20.21 (18.45) 33.48 (72.53) 20.63 (29.21) 28.32 (31.87) 21.56 (20.14) Logged earnings 2.88 (0.92) 2.74 (0.72) 3.04 (0.89) 2.73 (0.74) 2.99 (0.83) 2.78 (0.75) Hours worked 2239.59 (794.30) 1856.27 (753.50) 2234.42 (743.36) 1747.60 (722.98 2172.54 (735) 1735.30 (720.19) Human cap (%) < HS 15.01 4.06 11.43 11.7 5.59 6.06 HS grad 20.21 13.2 17.88 22.79 13.17 15.57 Some college 18.84 25.38 21.43 24.18 19.37 20.86 College grad 26.03 38.07 24.96 24.79 33.48 35.89 > College grad 19.91 19.29 24.30 16.55 28.39 21.63 Years of schooling (M.SD) 12.50 (6.36) 13.27 (6.10) 13.20 (6.37) 12.26 (6.32) 14.36 (6.08) 13.39 (6.34) Years of exp (M.SD) 19.09 (10.73) 18.58 (11.20) 24.08 (11.07) 25.25 (11.88) 22.81 (10.56) 24.32 (11.57) Experience2 (M.SD) 479.47 (522.09) 469.92 (525.45) 702.25 (586.65) 778.66 (643.40) 631.59 (532.27) 725.41 (607.53) Citizenship (%) Nat. citizen 58.9 65.31 72.37 79.66 59.73 64.88 Non-citizen 41.1 34.69 27.63 20.34 40.27 35.12 Occupation (%) Management 59.28 51.72 58.57 47.92 60.6 50.72 Service 13.77 13.79 15.16 12.71 14.85 15.94 Farming 6.595.177.099.054.077.83 English (%)

None 0.73 1.07 1.07 0.52 0.36 0.0 al. et A.M. Kusow Only 13.33 21.44 22.16 15.51 22.69 0.3 Very well 57.45 48.24 47.39 57.96 49.93 0.3 Good 22.91 20.21 21.7 21.21 21.83 99.4 Poor 5.58 9.04 7.67 4.79 5.19 oieooi civmn mn rbImgat nteUA the... USA: the in Immigrants Arab Among Achievement Socioeconomic Table 2 (continued)

Work (M.SD) Arabian Peninsula Levant North African

Male Female Male Female Male Female n =653 n =197 n = 3630 n =1650 n =2772 n =1304

Location (%) Northeast 23.28 21.32 18.43 19.09 33.48 31.13 Midwest 28.33 18.78 28.48 27.58 12.66 11.2 West 21.13 25.89 29.42 33.58 21.57 25.08 South 27.26 34.01 23.66 19.76 32.29 32.59 Married 78.1 67.01 78.02 72.73 76.48 75.61 N 2772 1034 3630 1650 653 197

Numbers in parenthesis are standard deviations for non-dichotomous variables 119 120 Kusow A.M. et al.

Table 3 OLS regression of logged hourly earnings for Arab immigrants (ACS, 2001–2913)

Model 1 Model 2 Model 3

Beta (SE) Beta (SE) Beta (SE)

Region North Africa 0.10** (0.03) − 0.05 (0.03) − 0.03 (0.03) Levant 0.12*** (0.03) 0.05 (0.03) 0.03 (0.03) Gender Male 0.30*** (0.02) 0.22*** (0.02) 0.24*** (0.02) Education and occupation Years of exp 0.02*** (0.00) 0.01*** (0.00) Experience2 − 0.00*** (0.00) − 0.00*** (0.00) High school grad 0.19*** (0.03) 0.12*** (0.03) Some college 0.39*** (0.03) 0.21*** (0.03) Bachelors degree 0.88*** (0.03) 0.60*** (0.03) > Bachelor’s degree 1.97*** (0.03) 1.46*** (0.03) Management 0.21*** (0.02) 0.18*** (0.02) Service 0.12** (0.04) 0.11** (0.04) Social and assimilation factors Married 0.05** (0.02) West 0.12*** (0.02) Northeast 0.07** (0.02) Midwest − 0.03 (0.02) Naturalized 0.13*** (0.02) Speak no English − 0.22** (0.10) Speak good English − 0.21*** (0.02) Speak poor English − 0.27*** (0.03) Intercept 13.14*** (0.03) 5.67*** (0.05) 6.54*** (0.05) Degrees of freedom 3 11 19 R2 0.021173 0.201493 0.228568

*p < 0.05, **p < 0.01, ***p <0.001

$4223. Notably, only $739 separates the highest female earning group (those from the Levant) and the middle earning group (those from North Africa). Another key pattern shown in Table 2 is in educational differences between men and women. Notwithstanding the fact that women of all groups report much lower mean averages in income, the figures in education show that, among these Arab immigrants, females report educational levels that are either equal to or higher than their male counterparts. For instance, looking at the percentages of those with some college education and those who are college graduates, we find female immigrants from the Arabian Peninsula have significantly higher educational achievement than their male counterparts, 25 and 38%, respectively, versus 18 and 26%, respectively. We find a similar pattern among immigrants from North Africa, among whom women show a slightly higher percentage of college gradu- ation than their male counterparts, about 36 versus 33%. In contrast, both female and male immigrants from the Levant report similar educational achievement levels, roughly 24%. That Arab immigrant women report educational levels that are equal to or higher than their male counterparts runs counter to the dominant belief that Arab women are discriminated against as part of Arab culture (Read 2003). Clearly, Arab women are given educational opportunities. Socioeconomic Achievement Among Arab Immigrants in the USA: the... 121

There is also significant variability across the three regions in the other socioeconomic variables. Across all regions, female immigrants are more likely to be naturalized citizens than men (64.88, 79.66, and 65.33% female immigrants from North Africa, Levant, and the Arabian Peninsula, respectively, vs. 59.73, 72.37, and 58.9%, respectively, among their male counterparts). One reason for the higher levels of citizenship acquisition among women is likely related to the fact that women often migrate because of family and/or marriage ties (Diner 1983;Lee2013). In other words, women’s migration decision to the USA is less likely to be made for temporary aims and instead is made with the intent to stay (Pedraza 1991), hence encouraging citizenship. In contrast to citizenship acquisition, the pattern for occupation favors men across all regions. About 59% of the male immigrants from the Arab Peninsula are engaged in management occupations as opposed to 51% for women; 58 versus 47% in the case of immigrants from the Levant, and 60 versus 50% for those from North Africa. Data on marital status show that men from the Arabian Peninsula and those from the Levant are more likely to be married than their female counterparts. Unlike immigrants from the Arabian Peninsula and the Levant, female immigrants from North Africa are as equally likely to be married as their male counterparts. English proficiency is an important variable in structuring socioeconomic achievement. The data show variability in English proficiency across the three source regions and by gender. Female immigrants from the Arab Peninsula are more likely to report poor English proficiency than those from the Levant and North Africa. Taken together, the key conclusion that can be derived from the data presented in Table 2 is that while there were slight variations in earnings across immigrants from the three regions, women earned substantially lower than their male counterparts across all regions. The same pattern is shown for occupation, with men more likely to be in management occupations than their female counterparts across all regions. This is despite the fact that female immigrants report the same educational achievement or higher than their male counterparts across all regions and report higher citizenship acquisition levels.

Multivariate Analysis

Table 3 presents OLS regression results for three models that examine logged hourly earnings for Arab immigrants in the USA. We first regress our key independent variables, region of origin and gender, on hourly earnings (model 1). We present this model to set the baseline difference in earnings between the three regional groups and between men and women before controlling for variables pertaining to human capital and social assimilation factors (models 2 and 3). More specifically, model 1 tests the effect of region of origin and gender on hourly earnings, and model 2 adds human capital variables—years of experience, educational attainment, and occupation. Finally, model 3 adds social and assimilation variables: marital status, region of residence in the USA, citizenship status, and ability to speak English. As shown in Table 3, results from model 1 show that both region of origin and gender are significant factors in structuring variation in earnings among Arab immigrants in the USA. The hourly earnings of immigrants from North Africa and the Levant are significantly higher than their counterparts from the Arabian Peninsula. Estimates for model 1 indicate that immigrants from North Africa had hourly earnings 122 Kusow A.M. et al. that are 10% higher than those of immigrants from the Arabian Peninsula. Likewise, immigrants from the Levant had hourly earnings that are 12% higher than their counterparts from the Arabian Peninsula. More important, results from model 1 also show that immigrant men had earnings that are significantly higher than those of their female counterparts. Model 1 shows that, on average, women earned about 30% less than men. Note that the estimate for men is highly significant (p <0.0001). In model 2 we add the human capital variables: educational attainment, occupation, and labor market experience. The addition of these new variables represents some improvement in model fit (from R2 = 0.02 for model 1 to R2 = 0.20 for model 2). Results for model 2 reveal the addition of the human capital variables completely eliminates the effects of region of origin. While both region of origin estimates (North Africa and Levant) become negative in model 2, neither is statistically significant. That is to say, net of human capital variables, model 2 shows no statistically significant difference in the earnings of Arab immigrants from North Africa and the Levant, compared with those from the Arabian Peninsula. By contrast, gender remains highly significant in model 2. The addition of the human capital variables in model 2 reduces the impact of gender slightly, from an estimate of 0.30 to 0.23. In other words, the addition of the human capital variables only reduces the difference in earnings by about 7% (from about 30 to about 23%). Stated differently, model 2 shows that, net of the human capital variables, women still earn about 23% less than their male counterparts. Moreover, in model 2, all human capital variables: experience, educational attainment, and occupation are significant. Model 3 adds controls for social and assimilation variables—marital status, citizenship status, region of US residency, and English language proficiency. Here again, the significance of gender in structuring hourly earnings persists, while that of region of origins remains statistically insignificant. Note that the estimate for men remains highly significant (p < 0.0001). In fact, the estimate for men goes up slightly. After controlling for all of our independent variables in model 3, results show that the earnings of women are about 24% less than men.5 To test whether the third region of origin comparison (North Africa vs. Levant) also remains insignificant, we ran model 3 again with the Levant as the reference category. This analysis revealed a difference in earnings between Arab immigrants from North Africa and the Levant that remains statistically significant (p < 0.0001). In other words, net of the effects of all of our independent variables, Arab immigrants from North Africa had earnings that were significantly lower than those from the Levant. Arab immigrants from North Africa earned about 7% less than immigrants from the Levant. Further tests, however, revealed the North Africa-Levant earnings difference is only significant among males. Running separate models for males and females, results of model 3 show no statistically significant earnings difference between Arab immigrants from North Africa and the Levant. By contrast, among males, there is a statistically significant difference in earnings between Arab immigrants from North Africa and the

5 We ran separate within-region analysis (results not reported in Table 3), and those analyses reveal similar findings. The within region results of model 3 show females from North Africa, the Levant, and the Arabian Peninsula to have earned about 17, 29, and 29%, respectively, less than their male counterparts (For north Africa, B = 0.173, STD = 0.024, p < 0.0001; for the Levant, B = 0.2894, STD = 0.023, p < 0.0001; for the Arabian Peninsula and Gulf, B = 0.287, STD = 0.065, p < 0.0001). Socioeconomic Achievement Among Arab Immigrants in the USA: the... 123

Levant (p < 0.0001). Stated differently, the difference in earnings between Arab immigrants from North Africa and the Levant existed only among males, with Arab immigrants from North Africa earning about 9% less than immigrants from the Levant. No such statistically significant difference existed among females. We plot the unad- justed means obtained from the descriptive analysis (Table 2) to illustrate that region matters for men, but not women (see Fig. 1). Results of model 3 also indicate that married individuals earned significantly higher incomes than unmarried individuals. Arab immigrants residing in the West and Northeast reported earnings that were significantly higher than those reported by immigrants who resided in the South. However, immigrants residing in the Midwest reported lower hourly earnings than immigrants residing in the South. Arab immigrants who were naturalized citizens reported significantly higher hourly earnings than non-naturalized citizens. Finally, the ability to speak English is a significant predictor of earnings. Immigrants who spoke no English, poor English, and Bgood^ English reported earnings that are significantly lower than those that reported speaking BEnglish only^ or Bvery well.^ This finding is consistent with the literature in that immigrants who come from regions where English is spoken are more likely earn more than those come from regions where English is not spoken (Kusow 2007). Taken together, the multivariate analysis results revealed three key findings. First, net of earnings-related measures, Arab immigrants from the Arabian Peninsula earned about the same as Arab immigrants from North Africa and the Levant. Second, immigrants from North Africa earned significantly less than those from the Levant, but this was only true for males. And third, gender, net of other variables such as education and occupation, is a powerful predictor of earnings. On average, the earnings of men were about 24% higher than that of their female counterparts. Further tests reveal the male-female earnings difference remains significant within regions. We discuss implications of these findings in the section to immediately follow.

40 *** 35 $sgninraEylruoH 30

25 Arabian 20 Levant 15 North Africa 10

0 Men Women Gender

Fig. 1 Region matters for men’s earnings 124 Kusow A.M. et al.

Discussion

We set out to examine the impact of region of origin and gender on earnings among Arab immigrants in the USA from three different sending regions. We specifically asked whether or not logged hourly earnings of Arab immigrants significantly vary by region of origin and gender. An important finding of our study is the seemingly intersectional ways in which region of origin and gender structure the nature of earnings among Arab immigrants in the USA. Unlike earlier findings from other immigrant groups (Kusow et al. 2016), region of origin does not appear to significantly affect earnings in our study. This finding may reflect that the act of immigration has a leveling effect. In other words, though immigrants originate from diverse nations that vary in terms of economic resources, those who immigrate and stay to work may have similar needs and face similar opportunities and constraints in the USA simply because they are an immigrant. Though regional differences are minimal, results show that gender is a far more significant factor in structuring variation in earnings among Arab immigrants in two important ways. First, we find that male immigrants from North Africa earn significantly less than those from the Levant. This shows that region of origin is important, but only when we examine it in terms of gender. One reason for this discrepancy may be that male immigrants from the Levant have a long history of entrepreneurship (Abdulrahim 2009;David2000;Sengstock1982), providing more opportunities for economic advancement. Second, and more important, we find that men in general earn significantly more than women. This shows that gender is a powerful predictor of earnings among Arab immigrants in the USA despite differences in region of origin. In other words, Arab immigrant women’s experiences mirror that of other women in the USA in that they earn less than men (Institute for Women’s Policy Research 2015). The fact that Arab immigrant men in general earn significantly higher than their female counterparts is less a reflection of how Arab women are treated, and more of a statement on how women, in general, are treated in the American labor market. A pattern found in the USA since 2000 is that women make between 75 and 80 cents for every dollar made by a man (Proctor et al. 2016). This gender differential holds true for both Arab (Read and Oselin 2008) and other ethnic immigrant women (Read and Cohen 2007) even if they have educational levels that are equal or even higher than their male counterparts. In a national survey examining employment patterns among 12 ethnic groups of American women, Read and Cohen (2007) found consistent evidence between high educational levels and low employment among Arab, Iranian, Korean, and Asian Indian women. Our findings advance the literature in that up to now, most studies have concentrated on Arab immigrants as a group or one gender or the other alone. No study we are aware of has so far investigated the impact of region of origin and gender on economic achievement among Arab immigrants in the USA simultaneously. The combination of region of origin and gender comparison moves us away from treating all immigrants from the Arab world as an undifferentiated body while illuminating comparison between men and women. This is important because (1) Arab immigrants come from different social, ethnic, religious, and colonial historical backgrounds and (2) gender is a key element of immigrant incorporation. Findings suggest that educational attainment may not translate into economic benefits for Arab immigrant women. These findings Socioeconomic Achievement Among Arab Immigrants in the USA: the... 125 show that though Arab immigrant women appear advantaged in terms of education (Read 2004), this human capital does not translate into real earnings.

Limitations and Future Directions

Though the American Community Survey (ACS) Study data represent the only population-level data available for Arab Americans, it is still somewhat limiting. For example, given the fact that some segments of the Arab immigrant population lives and works inside ethnic enclave economies, there exists the possibility of off-the-books employment which may not be captured by the self-reporting nature of the of the American Community Surveys. Consequently, a more comprehensive articulation of socioeconomic variation among Arab immigrants across different regions of origin obviously requires more robust data than we provide here. The Census data from which our analysis is derived do not have enough properly designed variables to fully address the questions at hand. Such an endeavor requires both historical and contextual variables that can address the different social and political contexts that inform emigration from Arab countries and the context of reception in host communities. Addi- tionally, it may be that examining where the immigrant completed education (Bratsberg and Ragan 2002) as well as reasons for immigration matter, measures we did not have. Furthermore, unpacking the Arab category even more may yield additional insights; country of origin may show more nuance patterns. Finally, better understanding of Arab American incorporation would be achieved if we could carry out a population based study that compared US-born Arab Americans to their immigrant counterparts. In sum, this study advances understandings of region of origin and gender patterns among Arab immigrants. Regional differences mattered most in the context of gender. Similar to general patterns in the USA, women earn less than men overall, suggesting that gender experiences concerning earnings among Arab immigrants are similar to the US population as a whole. Future research would benefit from a more in-depth examination of gender and socioeconomic activities to gain a better understanding of immigrant incorporation patterns.

References

Abdelhady, D. (2014). The sociopolitical history of Arabs in the United States: assimilation, ethnicity, and global citizenship. In S. C. Nassar-McMillan, K. J. Ajrouch, & J. Hakin-Larson (Eds.), Biopsychosocial perspectives on Arab Americans: culture, development, and health. New York: Springer Publishing. Abdulrahim, S. (2009). Pathways to social mobility Lebanese immigrants in Detroit and small business enterprise. Palma Journal, 11(1), 163–179. Abraham, N. (1994). Anti-Arab racism and violence in the United States. In E. McCarus (Ed.), The development of Arab American identity (pp. 155–214). Ann Arbor: University of Michigan Press. Abu-Laban, S. M. (1991). Family and religion among Muslim immigrants and their descendants. In E. W. Waugh, S. M. Abu-Laban, & R. B. Qureshi (Eds.), Muslim families in North America (pp. 6–31). Edmonton: University of Alberta Press. Ajrouch, K. J. (2004). Gender, race, and symbolic boundaries: contested spaces of identity among Arab American adolescents. Sociological Perspectives, 47(4), 371–391. Ajrouch, K. J., & Jamal, A. (2007). Assimilating to a white identity: the case of Arab Americans. International Migration Review, 41(4), 860–879. 126 Kusow A.M. et al.

Bashi, V., & McDaniel, A. (1997). A theory of immigration and racial stratification. Journal of Black Studies, 27(5), 668–682. Borch, C., & Corra, M. K. (2010). Differences in earnings among black and white African immigrants in the United States, 1980–2000: a cross-sectional and temporal analysis. Sociological Perspectives, 53(4), 573– 592. Bratsberg, B., & Ragan Jr., J. F. (2002). The impact of host-country schooling on earnings: a study of male immigrants in the United States. Journal of Human Resources, 37(1), 63–105. Butcher, K. (1994). Black Immigrants in the United States: A Comparison with Native Blacks and Other Immigrants. Industrial and Labor Relations Review 47:265-84. Cainkar, L. A. (2009). Homeland insecurity: the Arab American and Muslim American experience after 9/11. New York: Russell Sage Foundation. Corra, M. K., & Borch, C. (2014). Socioeconomic differences among blacks in America: over time trends. Race and Social Problems, 6(2), 103–119. Corra, M. K., & Kimuna, S. R. (2009). Double jeopardy? Female African and Caribbean immigrants in the United States. Journal of Ethnic and Migration Studies, 35(6), 1015–1035. Dallo, F. J., Snih, S. A., & Ajrouch, K. J. (2009). Prevalence of disability among US- and foreign-born Arab Americans: results from the 2000 US Census. Gerontology, 55(2), 153–161. David, G. (2000). Behind the bulletproof glass: Iraqi Chaldean store ownership in metropolitan Detroit. In N. Abraham & A. Shryock (Eds.), Arab Detroit: from margin to mainstream (pp. 151–178). Detroit: Wayne State University Press. Diner, H. R. (1983). Erin’s daughters in America: Irish immigrant women in the nineteenth century. Baltimore: Johns Hopkins University Press. Dodoo, F. Nii-Amoo. 1991a. Earnings Differences Among Blacks in America. Social Science Research 20:93–108. ___. 1991b. “mmigrant and Native Black Workers’ Labor Force Participation in the United States. National Journal of Sociology 5:1–17. ___. 1991c. “Blacks and Earnings in New York State. Sociological Spectrum 11:203–12. ___. 1991d. Minority Immigrants in the United States: Earnings Attributes and Economic Success. Canadian Studies in Population 18:42–55. Dodoo, F Nii-Amoo. 1997. Assimilation differences among Africans in America. Social Forces 76:527–46. Dodoo, F. N.-A., & Baffour, T. K. (2002). Africans in the diaspora: black-white earnings differences among America’s Africans. Ethnic and Racial Studies, 25(6), 913–941. Espiritu, Y. L. (1992). Asian American panethnicity: bridging institutions and identities. Philadelphia: Temple University Press. Field-Hendrey, E., & Balkan, E. (1991). Earnings and assimilation of female immigrants. Applied Economics, 23(10), 1665–1672. Gualtieri, S. (2001). Becoming Bwhite^: race, religion and the foundations of Syrian/Lebanese ethnicity in the United States. Journal of American Ethnic History, 20(4), 29–58. Hondagneu-Sotelo, P. (1999). Introduction: gender and contemporary U.S. immigration. American Behavioral Scientist, 42(4), 565–576. Institute for Women’s Policy Research (2015). https://iwpr.org/publications/the-status-of-women-in-the-states- 2015-full-report Retreived October 11, 2017. Kalmijn, M. (1996). The socioeconomic assimilation of Caribbean American blacks. Social Forces, 74(3), 911–930. Kanaiaupuni, S. M. (2000). Reframing the migration question: an analysis of men, women, and gender in Mexico. Social Forces, 74(3), 1311–1347. Kramer, M. (1993). Arab nationalism: mistaken identity. Daedalus, 122(3), 171–206. Kusow, A. (2007). Africa: east. In M. Waters & R. Ueda (Eds.), The new Americans: a guide to immigration since 1965 (pp. 295–306). Cambridge: Harvard University Press. Kusow, A. M., Kimuna, S. R., & Corra, M. (2016). Socioeconomic diversity among African immigrants in the United States: an intra-African immigrant comparison. Journal of International Migration and Integration, 17,115–130. Lee, C. (2013). Fictive kinship: family reunification and the meaning of race and nation in American migration. New York: Russell Sage Foundation. Model, S. (1991). Caribbean immigrants: a black success story? International Migration Review, 25(2), 248– 276. Model, S. (1995). West Indian prosperity: fact or fiction? Social Problems, 42(4), 535–553. Model, S. (2008). West Indian immigrants: a black success story? New York: Russell Sage Foundation. Moghadam, V. M. (2004). Patriarchy in transition: women and the changing family in the Middle East. Journal of Comparative Family Studies, 35(2), 137–162. Socioeconomic Achievement Among Arab Immigrants in the USA: the... 127

Naber, N., & Jamal, A. (2008). Race and Arab Americans before and after 9/11: from invisible citizens to visible subjects. Syracuse: Syracuse University Press. Naff,A.(1985).Becoming American: the early Arab immigrant experience. Carbondale and Edwardsville: Southern Illinois University Press. Nasser-McMillan, S., Ajrouch, K. J., & Hakim-Larson, J. (Eds.). (2014). Biopsychosocial perspectives on Arab Americans: culture, development, and health. New York: Springer Publications. Nawyn, S. J. (2010). Gender and migration: integrating feminist theory into migration studies. Sociology Compass, 4(9), 749–765. Nigem, E. T. (1986). Arab Americans: migration, socioeconomic and demographic characteristics. International Migration Review, 20(3), 629–629. Orfaela, G. (1988). Before the flames: a quest for the history of Arab Americans. Austin: University of Texas Press. Pedraza, S. (1991). Women and migration: the social consequences of gender. Annual Review of Sociology, 17, 303–325. Pedraza-Bailey, S. (1985). Political and economic migrants in America: Cubans and Mexicans. Austin: University of Texas Press. Pintak, L. (2009). Border guards of the Bimagined^ Watan: Arab journalists and the new Arab consciousness. The Middle East Journal, 63(2), 191–212. Portes, A., & Zhou, M. (1993). The new second generation: segmented assimilation and its variants. The Annals of the American Academy of Political and Social Science, 530(1), 74–96. Proctor, B. D., Semega, J. L., & Kollar, M. A. (2016). Income and poverty in the United States: 2015. Washington, DC: United States Census Bureau September. Read, J.’n. G. (2003). The sources of gender role attitudes among Christian and Muslim Arab-American women. Sociology of Religion, 64(2), 207–222. Read, J.’n. G. (2004). Family, religion, and work among Arab American women. Journal of Marriage and Family, 66(4), 1042–1050. Read, J. N. G. (2013). Measuring Ethnicity with US Census Data: Implications for Mexicans and Arabs. Population Research and Policy Review, 32(4), 611–631. Read, J.’n. G., & Cohen, P. N. (2007). One size fits all? Explaining US-born and immigrant women’s employment across 12 ethnic groups. Social Forces, 85(4), 1713–1734. Read, J.’n. G., & Emmerson, M. O. (2005). Racial context, black immigration and the U.S. black/white health disparity. Social Forces, 84(1), 181–199. Read, J.’n. G., & Oselin, S. (2008). Gender and the education-employment paradox in ethnic and religious contexts: the case of Arab Americans. American Sociological Review, 73(2), 296 –313. Roos, P. A. (1990). Review of revolving doors: sex segregation and women’s careers, by Jerry Jacobs. American Journal of Sociology, 95(5), 1315–1316. Salari, S. (2002). Invisible in aging research: Arab Americans, Middle Eastern immigrants, and Muslims in the United States. The Gerontologist, 42(5), 580–588. Samhan, H. H. (1999). Not quite white: race classification and the Arab-American experience. In M. W. Suleiman (Ed.), Arabs in America: building a new future (pp. 209–226). Philadelphia: Temple University Press. Samhan, H. H. (2014). Intra-ethnic diversity and religion. In S. Nasser-McMillan, K. J. Ajrouch, & J. Hakim- Larson (Eds.), Biopsychosocial perspectives on Arab Americans (pp. 45–65). New York: Springer. Schoeni, R. F. (1998). Labor market assimilation of immigrant women. Industrial & Labor Relations Review, 51(3), 483–504. Sengstock, M. C. (1982). The Chaldean Americans: changing conceptions of ethnic identity. New York: Center for Migration Studies. Sidani, Y. (2005). Women, work, and Islam in Arab societies. WomeninManagementReview,20(7), 498–512. Sirkeci, I. (2005). War in Iraq: environment of insecurity and international migration. International Migration, 43(4), 197–214. Suleiman, M. (1999). Arabs in America: building a new future. Philadelphia: Temple University Press. World Bank (2011). http://data.worldbank.org Retreived October 11, 2107 Zogby, J. (1990). Arab America today: a demographic profile of Arab Americans. Washington DC: Arab American Institute. Zsembik, B. A., & Fennell, D. (2005). Ethnic variation in health and the determinants of health among Latinos. Social Science & Medicine, 61(1), 53–63.